Objective: Bioelectrical impedance analysis (BIA) is a commonly used method for the evaluation of body composition. However, BIA estimations are subject to uncertainties. The aim of this systematic review was to explore the variability of empirical prediction equations used in BIA estimations and to evaluate the validity of BIA estimations in adult surgical and oncological patients. Subjects: Studies developing new empirical prediction equations and studies evaluating the validity of BIA estimations compared with a reference method were included. Only studies using BIA devices measuring the entire body were included. Studies that included patients with altered body composition or a disturbed fluid balance and studies written in languages other than English were excluded. To illustrate variability between equations, fixed normal reference values of resistance values were entered into the existing empirical prediction equations of the included studies and the results were plotted in figures. The validity was expressed by the difference in means between BIA estimates and the reference method, and relative difference in %. Results: Substantial variability between equations for groups (including men and women) was found for total body water (TBW) and fat free mass (FFM). The gender-specific existing general equations assume less variability for TBW and FFM. BIA mainly underestimated TBW (range relative difference -18.8% to +7.2%) and FFM (range relative differences -15.2% to +3.8%). Estimates of the fat mass (FM) demonstrated large variability (range relative difference -15.7 to +43.1%). Conclusions: Application of equations validated in healthy subjects to predict body composition performs less well in oncologic and surgical patients. We suggest that BIA estimations, irrespective of the device, can only be useful when performed longitudinally and under the same standard conditions.
Bibliographical notePublisher Copyright:
© 2015 Macmillan Publishers Limited.